Endometrial adenocarcinoma is the most common malignancy of the female genital tract, and occurs at elevated frequency in women who have premalignant endometrial lesions or are exposed to estrogens for protracted intervals. Somatically acquired loss of function of the PTEN tumor suppressor gene is an initiating event in sporadic endometrial carcinogenesis, which characterizes two thirds of histologically evident precancers (EIN, or atypical endometrial hyperplasia) and their resultant endometrial cancers. Small numbers of PTEN-null histologically unremarkable endometrial glands, "latent precancers" can be found in the endometrium of almost half of normal women, indicating that progression efficiency rather than initiation, is the major barrier to carcinogenesis. We postulate that hormonal factors which modulate endometrial cancer risk act to cause expansion/persistence or involution/disappearance of pre-existing somatically acquired mutant endometrial clones. We will test this hypothesis by tracking the fate of PTEN-null endometrial clones in sequential biopsies of women undergoing hormonal replacement therapy known to increase (unopposed estrogen) or decrease (combined estrogen and progestin) risk of endometrial adenocarcinoma. Changing trends in the fraction of women with PTEN-null endometrial clones will be measured as a function of hormonal exposure type and duration (SA#1). In SA#2, specific PTEN mutations will be used as clonal markers to distinguish between persistence of individual clones over time and successive initiation of multiple short lived clones. SA#3 will correlate endometrial histopathology with clonal progression from a histologically latent to diagnosable stage. If our hypothesis is confirmed, this project will establish biomarker-detectable preclinical disease as a surrogate marker for risk surveillance, and a potential target for cancer preventative therapies.